Sp1 functions in a chromatin-dependent manner to augment human alpha-globin promoter activity.

The 5' flanking region of the human alpha-globin gene is highly G + C rich and contains multiple copies of the consensus sequence for the Sp1 binding site. We investigated the role of this G + C-rich region in augmenting alpha-globin promoter activity in the presence of the far-upstream alpha-globin enhancer, HS-40. We show that in transiently transfected erythroid cells, deletion of the alpha-globin G + C-rich 5' flanking region has no effect on alpha-globin promoter activity. However, upon stable integration into chromatin, deletion of this region causes a nearly 90% decrease in promoter activity compared with expression from an alpha-globin promoter retaining this region. These results suggest that the alpha-globin G + C-rich 5' flanking region augments alpha-globin promoter activity in a chromatin-dependent manner. We further show that this G + C-rich region is required for the activation of alpha-globin gene expression during erythroid differentiation. Finally, we show by both footprint analysis and functional assays that the ability of the G + C-rich region to increase alpha-globin promoter activity from a stably integrated alpha-globin gene is mediated by its multiple binding sites for the transcription factor Sp1.

Integration host factor suppresses promiscuous activation of the sigma 54-dependent promoter Pu of Pseudomonas putida.

In the presence of m-xylene, the Pu promoter of the TOL plasmid of Pseudomonas putida is activated by the prokaryotic enhancer-binding protein XylR. The intervening DNA segment between the upstream activating sequences (UASs) and those for RNA polymerase binding contains an integration host factor (IHF) attachment site that is required for full transcriptional activity. In the absence of IHF, the Pu promoter can be cross-activated by other members of the sigma 54-dependent family of regulatory proteins. Such illegitimate activation does not require the binding of the heterologous regulators to DNA and it is suppressed by bent DNA structures, either static or protein induced, between the promoter core elements (UAS and RNA polymerase recognition sequence). The role of IHF in some sigma 54 promoters is, therefore, not only a structural aid for assembling a correct promoter geometry but also that of an active suppressor (restrictor) of promiscuous activation by heterologous regulators for increased promoter specificity.

Stringent chemical and thermal regulation of recombinant gene expression by vaccinia virus vectors in mammalian cells.

We developed a stringently regulated expression system for mammalian cells that uses (i) the RNA polymerase, phi 10 promoter, and T phi transcriptional terminator of bacteriophage T7; (ii) the lac repressor, lac operator, rho-independent transcriptional terminators and the gpt gene of Escherichia coli; (iii) the RNA translational enhancer of encephalomyocarditis virus; and (iv) the genetic background of vaccinia virus. In cells infected with the recombinant vaccinia virus, reporter beta-galactosidase synthesis was not detected in the absence of inducer. An induction of at least 10,000- to 20,000-fold occurred upon addition of isopropyl beta-D-thiogalactopyranoside or by temperature elevation from 30 to 37 degrees C using a temperature-sensitive lac repressor. Regulated synthesis of the secreted and highly glycosylated human immunodeficiency virus 1 envelope protein gp120 was also demonstrated. Yields of both proteins were approximately 2 mg per 10(8) cells in 24 hr. Plasmid transfer vectors for cloning and expression of complete or incomplete open reading frames in recombinant vaccinia viruses are described.

Green fluorescent protein as a vital marker and reporter of gene expression in Drosophila.

We have used the green fluorescent protein (GFP) from the jellyfish Aequorea victoria as a vital marker/reporter in Drosophila melanogaster. Transgenic flies were generated in which GFP was expressed under the transcriptional control of the yeast upstream activating sequence that is recognized by GAL4. These flies were crossed to several GAL4 enhancer trap lines, and expression of GFP was monitored in a variety of tissues during development using confocal microscopy. Here, we show that GFP could be detected in freshly dissected ovaries, imaginal discs, and the larval nervous system without prior fixation or the addition of substrates or antibodies. We also show that expression of GFP could be monitored in intact living embryos and larvae and in cultured egg chambers, allowing us to visualize dynamic changes in gene expression during real time.

Enhancers increase the probability but not the level of gene expression.

We have studied enhancer function in transient and stable expression assays in mammalian cells by using systems that distinguish expressing from nonexpressing cells. When expression is studied in this way, enhancers are found to increase the probability of a construct being active but not the level of expression per template. In stably integrated constructs, large differences in expression level are observed but these are not related to the presence of an enhancer. Together with earlier studies, these results suggest that enhancers act to affect a binary (on/off) switch in transcriptional activity. Although this idea challenges the widely accepted model of enhancer activity, it is consistent with much, if not all, experimental evidence on this subject. We hypothesize that enhancers act to increase the probability of forming a stably active template. When randomly integrated into the genome, enhancers may affect a metastable state of repression/activity, permitting expression in regions that would not permit activity of an isolated promoter.

Delineation of a slow-twitch-myofiber-specific transcriptional element by using in vivo somatic gene transfer.

Contractile proteins are encoded by multigene families, most of whose members are differentially expressed in fast- versus slow-twitch myofibers. This fiber-type-specific gene regulation occurs by unknown mechanisms and does not occur within cultured myocytes. We have developed a transient, whole-animal assay using somatic gene transfer to study this phenomenon and have identified a fiber-type-specific regulatory element within the promoter region of a slow myofiber-specific gene. A plasmid-borne luciferase reporter gene fused to various muscle-specific contractile gene promoters was differentially expressed when injected into slow- versus fast-twitch rat muscle: the luciferase gene was preferentially expressed in slow muscle when fused to a slow troponin I promoter, and conversely, was preferentially expressed in fast muscle when fused to a fast troponin C promoter. In contrast, the luciferase gene was equally well expressed by both muscle types when fused to a nonfiber-type-specific skeletal actin promoter. Deletion analysis of the troponin I promoter region revealed that a 157-bp enhancer conferred slow-muscle-preferential activity upon a minimal thymidine kinase promoter. Transgenic analysis confirmed the role of this enhancer in restricting gene expression to slow-twitch myofibers. Hence, somatic gene transfer may be used to rapidly define elements that direct myofiber-type-specific gene expression prior to the generation of transgenic mice.

Scavenger receptor A gene regulatory elements target gene expression to macrophages and to foam cells of atherosclerotic lesions.

Transcription of the macrophage scavenger receptor A gene is markedly upregulated during monocyte to macrophage differentiation. In these studies, we demonstrate that 291 bp of the proximal scavenger receptor promoter, in concert with a 400-bp upstream enhancer element, is sufficient to direct macrophage-specific expression of a human growth hormone reporter in transgenic mice. These regulatory elements, which contain binding sites for PU.1, AP-1, and cooperating ets-domain transcription factors, are also sufficient to mediate regulation of transgene expression during the in vitro differentiation of bone marrow progenitor cells in response to macrophage colony-stimulating factor. Mutation of the PU.1 binding site within the scavenger receptor promoter severely impairs transgene expression, consistent with a crucial role of PU.1 in regulating the expression of the scavenger receptor gene. The ability of the scavenger receptor promoter and enhancer to target gene expression to macrophages in vivo, including foam cells of atherosclerotic lesions, suggests that these regulatory elements will be of general utility in the study of macrophage differentiation and function by permitting specific modifications of macrophage gene expression.

Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension.

Hypoxia-inducible factor 1 (HIF-1) is found in mammalian cells cultured under reduced O2 tension and is necessary for transcriptional activation mediated by the erythropoietin gene enhancer in hypoxic cells. We show that both HIF-1 subunits are basic-helix-loop-helix proteins containing a PAS domain, defined by its presence in the Drosophila Per and Sim proteins and in the mammalian ARNT and AHR proteins. HIF-1 alpha is most closely related to Sim. HIF-1 beta is a series of ARNT gene products, which can thus heterodimerize with either HIF-1 alpha or AHR. HIF-1 alpha and HIF-1 beta (ARNT) RNA and protein levels were induced in cells exposed to 1% O2 and decayed rapidly upon return of the cells to 20% O2, consistent with the role of HIF-1 as a mediator of transcriptional responses to hypoxia.

Combinatorial association and abundance of components of interferon-stimulated gene factor 3 dictate the selectivity of interferon responses.

Genes containing the interferon-stimulated response element (ISRE) enhancer have been characterized as transcriptionally responsive primarily to type I interferons (IFN alpha/beta). Induction is due to activation of a multimeric transcription factor, interferon-stimulated gene factor 3 (ISGF3), which is activated by IFN alpha/beta but not by IFN gamma. We found that ISRE-containing genes were induced by IFN gamma as well as by IFN alpha in Vero cells. The IFN gamma response was dependent on the ISRE and was accentuated by preexposure of cells to IFN alpha, a treatment that increases the abundance of ISGF3 components. Overexpression of ISGF3 polypeptides showed that the IFN gamma response depended on the DNA-binding protein ISGF3 gamma (p48) as well as on the 91-kDa protein STAT91 (Stat1 alpha). The transcriptional response to IFN alpha required the 113-kDa protein STAT113 (Stat2) in addition to STAT91 and p48. Mutant fibrosarcoma cells deficient in each component of ISGF3 were used to confirm that IFN gamma induction of an ISRE reporter required p48 and STAT91, but not STAT113. A complex containing p48 and phosphorylated STAT91 but lacking STAT113 bound the ISRE in vitro. IFN gamma-induced activation of this complex, preferentially formed at high concentrations of p48 and STAT91, may explain some of the overlapping responses to IFN alpha and IFN gamma.

Correlation of peptide specificity and IgG subclass with pathogenic and nonpathogenic autoantibodies in pemphigus vulgaris: a model for autoimmunity.

Pemphigus vulgaris (PV) is a rare, potentially fatal, autoimmune disease that affects the skin and mucous membranes. The PV antigen (PVA) has been characterized as desmoglein 3. PV patients carry HLA-DR4- or HLA-DR6-bearing extended haplotypes. We recently demonstrated that patients with active disease have high titers of PV autoantibodies of the IgG1 and IgG4 subclasses. Patients in remission, healthy unaffected relatives, and some MHC-matched normal individuals have low levels of PV autoantibodies, which are IgG1 only. Furthermore, intraperitoneal injection of IgG from patients with active disease caused clinical disease in mice, but IgG from patients in remission, healthy relatives, or MHC-matched normal individuals did not. We prepared 12 peptides of 30 amino acids each (peptides Bos 1-12) spanning the extracellular domain of PVA. Patients with active disease recognize peptides Bos 1 and Bos 6 with high titers of IgG1 and IgG4 autoantibodies. Patients in remission have IgG1 autoantibodies to peptide Bos 1 only, in statistically significantly lower titers (P < 0.01). They no longer have IgG4 subclass autoantibodies to peptide Bos 6. Healthy relatives and normal unrelated individuals have low levels of only IgG1 autoantibodies that recognize only Bos 1. In vitro studies indicate that Bos 6-specific IgG and, to a lesser extent, Bos 1-specific IgG can cause acantholysis. Our data suggest that Bos 6-specific IgG4 is probably the main acantholytic autoantibody, while Bos 1-specific IgG4 may act as a facilitator or enhancer of the process. In this study we illustrate some of the paradigms that demonstrate the interactions between the MHC, subclass of autoantibodies, and peptide specificities of the autoantibodies in the autoimmune process. Thus, PV provides an important model to study the pathogenesis of autoimmunity.

Mammary tumor suppression by transforming growth factor beta 1 transgene expression.

In cell culture, type alpha transforming growth factor (TGF-alpha) stimulates epithelial cell growth, whereas TGF-beta 1 overrides this stimulatory effect and is growth inhibitory. Transgenic mice that overexpress TGF-alpha under control of the mouse mammary tumor virus (MMTV) promoter/enhancer exhibit mammary ductal hyperplasia and stochastic development of mammary carcinomas, a process that can be accelerated by administration of the chemical carcinogen 7,12-dimethylbenz[a]anthracene. MMTV-TGF-beta 1 transgenic mice display mammary ductal hypoplasia and do not develop mammary tumors. We report that in crossbreeding experiments involving the production of mice carrying both the MMTV-TGF-beta 1 and MMTV-TGF-alpha transgenes, there is marked suppression of mammary tumor formation and that MMTV-TGF-beta 1 transgenic mice are resistant to 7,12-dimethylbenz[a]anthracene-induced mammary tumor formation. These data demonstrate that overexpression of TGF-beta 1 in vivo can markedly suppress mammary tumor development.

Pax-6 and lens-specific transcription of the chicken delta 1-crystallin gene.

The abundance of delta-crystallin in the chicken eye lens provides an advantageous marker for tissue-specific gene expression during cellular differentiation. The lens-specific expression of the delta 1-crystallin gene is governed by an enhancer in the third intron, which binds a positive (delta EF2) and negative (delta EF1) factor in its core region. Here we show by DNase I footprinting, electrophoretic mobility-shift assays, and cotransfection experiments with the delta 1-promoter/enhancer fused to the chloramphenicol acetyltransferase reporter gene that the delta 1-crystallin enhancer has two adjacent functional Pax-6 binding sites. We also demonstrate by DNase I footprinting that the delta EF1 site can bind the transcription factor USF, raising the possibility that USF may cooperate with Pax-6 in activation of the chicken delta 1- and alpha A-crystallin genes. These data, coupled with our recent demonstration that Pax-6 activates the alpha A-crystallin gene, suggest that Pax-6 may have been used extensively throughout evolution to recruit and express crystallin genes in the lens.

Regulatory region in choline acetyltransferase gene directs developmental and tissue-specific expression in transgenic mice.

Acetylcholine, one of the main neurotransmitters in the nervous system, is synthesized by the enzyme choline acetyltransferase (ChAT; acetyl-CoA:choline O-acetyltransferase, EC 2.3.1.6). The molecular mechanisms controlling the establishment, maintenance, and plasticity of the cholinergic phenotype in vivo are largely unknown. A previous report showed that a 3800-bp, but not a 1450-bp, 5' flanking segment from the rat ChAT gene promoter directed cell type-specific expression of a reporter gene in cholinergic cells in vitro. Now we have characterized a distal regulatory region of the ChAT gene that confers cholinergic specificity on a heterologous downstream promoter in a cholinergic cell line and in transgenic mice. A 2342-bp segment from the 5' flanking region of the ChAT gene behaved as an enhancer in cholinergic cells but as a repressor in noncholinergic cells in an orientation-independent manner. Combined with a heterologous basal promoter, this fragment targeted transgene expression to several cholinergic regions of the central nervous system of transgenic mice, including basal forebrain, cortex, pons, and spinal cord. In eight independent transgenic lines, the pattern of transgene expression paralleled qualitatively and quantitatively that displayed by endogenous ChAT mRNA in various regions of the rat central nervous system. In the lumbar enlargement of the spinal cord, 85-90% of the transgene expression was targeted to the ventral part of the cord, where cholinergic alpha-motor neurons are located. Transgene expression in the spinal cord was developmentally regulated and responded to nerve injury in a similar way as the endogenous ChAT gene, indicating that the 2342-bp regulatory sequence contains elements controlling the plasticity of the cholinergic phenotype in developing and injured neurons.

Two auxin-responsive domains interact positively to induce expression of the early indoleacetic acid-inducible gene PS-IAA4/5.

The plant growth hormone indole-3-acetic acid (IAA) transcriptionally activates expression of several genes in plants. We have previously identified a 164-bp promoter region (-318 to -154) in the PS-IAA4/5 gene that confers IAA inducibility. Linker-scanning mutagenesis across the region has identified two positive domains: domain A (48 bp; -203 to -156) and domain B (44 bp; -299 to -256), responsible for transcriptional activation of PS-IAA4/5 by IAA. Domain A contains the highly conserved sequence 5'-TGTCCCAT-3' found among various IAA-inducible genes and behaves as the major auxin-responsive element. Domain B functions as an enhancer element which may also contain a less efficient auxin-responsive element. The two domains act cooperatively to stimulate transcription; however, tetramerization of domain A or B compensates for the loss of A or B function. The two domains can also mediate IAA-induced transcription from the heterologous cauliflower mosaic virus 35S promoter (-73 to +1). In vivo competition experiments with icosamers of domain A or B show that the domains interact specifically and with different affinities to low abundance, positive transcription factor(s). A model for transcriptional activation of PS-IAA4/5 by IAA is discussed.

Isolation of a differentially regulated splicing isoform of human NF-E2.

The transcription factor NF-E2 (nuclear factor erythroid 2), interacting via DNA motifs within regulatory regions of several hematopoietic genes, is thought to mediate the enhancer activity of the globin locus control regions. By screening a human fetal liver cDNA library with probes derived from mouse NF-E2, we have isolated a splicing variant of the NF-E2 gene (fNF-E2) that differs in the 5' untranslated region from the previously reported cDNA (aNF-E2). The fNF-E2 isoform is transcribed from an alternative promoter located in the 3' end of the first intron and joined by alternative splicing to the second and third exons, which are shared by both RNA isoforms. Although the two forms produce the same protein, they are expressed in different ratios during development. fNF-E2 is more abundant in the fetal liver and less abundant in the adult bone marrow compared to the previously described form. Their distribution apparently follows the differential expression of fetal and adult hemoglobins.